Brain image analysis of ASD using large-scale deep learning models

Abstract

Autism Spectrum Disorder (ASD) is primarily diagnosed subjectively, which can be time-consuming. Therefore, an objective perspective is needed. This study proposes an analysis focusing on the structural features of the brain using MRI images. Moreover, in order to perform machine learning on an imbalanced dataset with unequal data points, sufficient consideration was given. While the required diagnostic accuracy has not been achieved yet, it has been confirmed that the analysis focusing on brain structure is effective.